The clearly defined genetic and phenotypic stages in the evolution of squamous tumor development on mouse skin provides an opportunity to evaluate mechanisms of carcinogenesis in a stage-specific manner. Focus on genetic changes frequently detected in human cancers such as mutations in ras or p53 and alterations in the epidermal growth factor receptor(EGFR)family is helpful in extrapolating results from the animal model to understand carcinogenesis in humans. Keratinocytes derived from EGFR deficient mice and transformed by the ras oncogene form very small tumors with a high growth fraction in vivo. Studies with these null keratinocytes in vitro reveal an accelerated migration rate and diminished attachment to extracellular matrix. These data are consistent with the rapid migration of BrdU labeled basal cells into the suprabasal compartment of EGFR deficient tumors and the premature cell cycle arrest as tumor cells enter the differentiating pool. Activation of the EGFR in squamous tumors causes tyrosine phosphorylation of PKC d and inactivation of its catalytic activity. Members of the Src family appear to be the proximal tyrosine kinases that phosphorylate PKC d. In tumor cells, Src and PKC d co-immunoprecipitate, and Src inhibitors prevent the association, inhibit PKC d tyrosine phosphorylation, and induce differentiation. The mitochondria of keratinocytes are a target for PKC d in normal keratinocytes. Activation of PKC d causes a decrease in mitochondrial membrane potential, activation of caspases and keratinocyte apoptosis. Inhibitors of mitochondrial electron transport can suppress this apoptotic response. Similarly, antioxidants also prevent apoptosis induced by PKC d. Squamous tumor cells of mouse and human origin are susceptible to killing by overexpressing and activating PKC d using an adenoviral vector. Overexpression of PKC a in transgenic epidermis produces a marked inflammatory response and apoptosis when activated by phorbol esters. PKC a activation upregulates the expression of a number of chemokines and alters the expression of keratinocyte differentiation markers. These affects of PKC a can be prevented by a dominant-negative Fos vector suggesting they are mediated by AP-1 dependent transcriptional activity.